Astrocytes play critical roles in the biology and diseases of the central nervous system, yet much remains to be learned about their function. Recent advances in the molecular biology of astrocyte-specific genes offer new opportunities for genetic manipulation of astrocyte function in vivo through transgenic techniques. Our goals are to understand how the glial- specific intermediate filament, glial fibrillary acidic protein (GFAP), contributes to the development and reactions to injury of astrocytes in the central nervous system. We propose to test the hypothesis that GFAP expression itself directly influences the ability of astrocytes to form processes in vivo, and to form gliotic scars following injury, by generation of two kinds of transgenic mice. First, we will force over- expression of GFAP in astrocytes by adding multiple copies of functional GFAP genes to the mouse germ-line. These mice will demonstrate whether up- regulation of GFAP itself can initiate astrocyte hypertrophy. Second, we will prevent GFAP expression by generation of mice specifically mutated at the GFAP locus. These mice will demonstrate whether GFAP is necessary for astrocyte development and gliosis. In each case the entire nervous system will be surveyed by light microscopy, but particular attention will be focussed on the cerebellum, spinal cord, optic nerve, and retina. Finally, we will examine whether changes in astrocyte expression of GFAP result in altered responses to injuries such as trauma. The proposed experiments promise further understanding of the molecular and cell biological properties of GFAP in astrocytes, and ultimately the functions of astrocytes in normal development and diseases of the central nervous system.